Stentor A new Computer-Aided Transcription software for French language.

Slides:

Advertisements

Similar presentations

LABORATOIRE DINFORMATIQUE CERI 339 Chemin des Meinajariès BP AVIGNON CEDEX 09 Tél (0) Fax (0)

Advertisements

1 Speech Sounds Introduction to Linguistics for Computational Linguists.

English only VS. L1 support Looking at English language acquisition of native Spanish speaking children.

Research & Development ICASSP' Analysis of Model Adaptation on Non-Native Speech for Multiple Accent Speech Recognition D. Jouvet & K. Bartkova France.

Acoustic Model Adaptation Based On Pronunciation Variability Analysis For Non-Native Speech Recognition Yoo Rhee Oh, Jae Sam Yoon, and Hong Kook Kim Dept.

A Syntactic Translation Memory Vincent Vandeghinste Centre for Computational Linguistics K.U.Leuven

: Recognition Speech Segmentation Speech activity detection Vowel detection Duration parameters extraction Intonation parameters extraction German Italian.

INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE PROCESSING NLP-AI IIIT-Hyderabad CIIL, Mysore ICON DECEMBER, 2003.

Development of Automatic Speech Recognition and Synthesis Technologies to Support Chinese Learners of English: The CUHK Experience Helen Meng, Wai-Kit.

PHONEXIA Can I have it in writing?. Discuss and share your answers to the following questions: 1.When you have English lessons listening to spoken English,

Inducing Information Extraction Systems for New Languages via Cross-Language Projection Ellen Riloff University of Utah Charles Schafer, David Yarowksy.

1 Noun Homograph Disambiguation Using Local Context in Large Text Corpora Marti A. Hearst Presented by: Heng Ji Mar. 29, 2004.

Resources Primary resources – Lexicons, structured vocabularies – Grammars (in widest sense) – Corpora – Treebanks Secondary resources – Designed for a.

Dept. of Computer Science & Engg. Indian Institute of Technology Kharagpur Part-of-Speech Tagging and Chunking with Maximum Entropy Model Sandipan Dandapat.

Near-Duplicate Detection by Instance-level Constrained Clustering Hui Yang, Jamie Callan Language Technologies Institute School of Computer Science Carnegie.

The LC-STAR project (IST ) Objectives: Track I (duration 2 years) Specification and creation of large word lists and lexica suited for flexible.

An innovative platform to allow translation and indexing of internet sites Localization World

Building High Quality Databases for Minority Languages such as Galician F. Campillo, D. Braga, A.B. Mourín, Carmen García-Mateo, P. Silva, M. Sales Dias,

Natural Language Understanding

® Automatic Scoring of Children's Read-Aloud Text Passages and Word Lists Klaus Zechner, John Sabatini and Lei Chen Educational Testing Service.

Katherine S. Holmes READ 7140 May 28, Georgia Writing Test – 5 th Grade GOAL: To assess the procedures to enhance statewide instruction in language.

Computational Methods to Vocalize Arabic Texts H. Safadi*, O. Al Dakkak** & N. Ghneim**

Lemmatization Tagging LELA /20 Lemmatization Basic form of annotation involving identification of underlying lemmas (lexemes) of the words in.

BTANT 129 w5 Introduction to corpus linguistics. BTANT 129 w5 Corpus The old school concept – A collection of texts especially if complete and self-contained:

Real-Time Speech Recognition Subtitling in Education Respeaking 2009 Dr Mike Wald University of Southampton.

T raining on Read&Write GOLD Dick Powers

Whither Linguistic Interpretation of Acoustic Pronunciation Variation Annika Hämäläinen, Yan Han, Lou Boves & Louis ten Bosch.

Speech and Language Processing

MONGOLIAN TAGSET and CORPUS TAGGING J.Purev and Ch. Odbayar CRLP Center for Research on Language Processing National University of Mongolia (NUM)

 Feature extractor  Mel-Frequency Cepstral Coefficients (MFCCs) Feature vectors.

Querying Across Languages: A Dictionary-Based Approach to Multilingual Information Retrieval Doctorate Course Web Information Retrieval Speaker Gaia Trecarichi.

Learner corpus analysis and error annotation Xiaofei Lu CALPER 2010 Summer Workshop July 13, 2010.

Experiments on Building Language Resources for Multi-Modal Dialogue Systems Goals identification of a methodology for adapting linguistic resources for.

NLP Language Models1 Language Models, LM Noisy Channel model Simple Markov Models Smoothing Statistical Language Models.

1 BILC SEMINAR 2009 Speech Recognition: Is It for Real? Tony Mirabito Defense Language Institute English Language Center (DLIELC) DLIELC.

Machine Translation  Machine translation is of one of the earliest uses of AI  Two approaches:  Traditional approach using grammars, rewrite rules,

Introduction to CL & NLP CMSC April 1, 2003.

Using Short-Answer Format Questions for an English Grammar Tutoring System Conceptualization & Research Planning Jonggun Gim.

Modeling Speech using POMDPs In this work we apply a new model, POMPD, in place of the traditional HMM to acoustically model the speech signal. We use.

1 Boostrapping language models for dialogue systems Karl Weilhammer, Matthew N Stuttle, Steve Young Presenter: Hsuan-Sheng Chiu.

ELIS-DSSP Sint-Pietersnieuwstraat 41 B-9000 Gent Recognition of foreign names spoken by native speakers Frederik Stouten & Jean-Pierre Martens Ghent University.

PSY270 Michaela Porubanova. Language  a system of communication using sounds or symbols that enables us to express our feelings, thoughts, ideas, and.

Recurrent neural network based language model Tom´aˇs Mikolov, Martin Karafia´t, Luka´sˇ Burget, Jan “Honza” Cˇernocky, Sanjeev Khudanpur INTERSPEECH 2010.

Yuya Akita , Tatsuya Kawahara

French / English Translation by Sharon Ulery. Purpose computational linguistics to serve students of French or English & those who only know one of these.

Auckland 2012Kilgarriff: NLP and Corpus Processing1 The contribution of NLP: corpus processing.

Latent Topic Modeling of Word Vicinity Information for Speech Recognition Kuan-Yu Chen, Hsuan-Sheng Chiu, Berlin Chen ICASSP 2010 Hao-Chin Chang Department.

LREC 2004, 26 May 2004, Lisbon 1 Multimodal Multilingual Resources in the Subtitling Process S.Piperidis, I.Demiros, P.Prokopidis, P.Vanroose, A. Hoethker,

Supertagging CMSC Natural Language Processing January 31, 2006.

Hybrid Method for Tagging Arabic Text Written By: Yamina Tlili-Guiassa University Badji Mokhtar Annaba, Algeria Presented By: Ahmed Bukhamsin.

1 Minimum Error Rate Training in Statistical Machine Translation Franz Josef Och Information Sciences Institute University of Southern California ACL 2003.

AUTONOMOUS REQUIREMENTS SPECIFICATION PROCESSING USING NATURAL LANGUAGE PROCESSING - Vivek Punjabi.

Word classes and part of speech tagging. Slide 1 Outline Why part of speech tagging? Word classes Tag sets and problem definition Automatic approaches.

St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences Recurrent Neural Network-based Language Modeling for an Automatic.

1 ICASSP Paper Survey Presenter: Chen Yi-Ting. 2 Improved Spoken Document Retrieval With Dynamic Key Term Lexicon and Probabilistic Latent Semantic Analysis.

Phone-Level Pronunciation Scoring and Assessment for Interactive Language Learning Speech Communication, 2000 Authors: S. M. Witt, S. J. Young Presenter:

Basics of Natural Language Processing Introduction to Computational Linguistics.

Maximum Entropy techniques for exploiting syntactic, semantic and collocational dependencies in Language Modeling Sanjeev Khudanpur, Jun Wu Center for.

A Simple English-to-Punjabi Translation System By : Shailendra Singh.

An Efficient Hindi-Urdu Transliteration System Nisar Ahmed PhD Scholar Department of Computer Science and Engineering, UET Lahore.

Arnar Thor Jensson Koji Iwano Sadaoki Furui Tokyo Institute of Technology Development of a Speech Recognition System For Icelandic Using Machine Translated.

G. Anushiya Rachel Project Officer

Audio to Score Alignment for Educational Software

Linguistic knowledge for Speech recognition

Speaker : chia hua Authors : Long Qin, Ming Sun, Alexander Rudnicky

Tagging and Statistically Translating Latin Sentences

Linguistic insecurity (I don't know correct way)

汉语连续语音识别年1月4日访北京工业大学 973 Project 2019/4/17 汉语连续语音识别年1月4日访北京工业大学郑方清华大学计算机科学与技术系语音实验室

University of Illinois System in HOO Text Correction Shared Task

Artificial Intelligence 2004 Speech & Natural Language Processing

Presentation transcript:

Stentor A new Computer-Aided Transcription software for French language

SténoMédia Young compagny : 6 months Pretty long history : 6 years and 2 years Association of professional and researchers  S. Badot : International stenotypist  Y. Estève : Researcher, Université du Mans  T. Spriet : Researcher, Université d'Avignon Stenotypy domain : a very good lab

French Stenotypy ''Grandjean'' method, based on words pronunciation High rate of homophony, about 1.8 long span syntactic constraints various application domains

Stenotypy vs Speech recognition ''acoustic variations'' due to typing errors ''acoustic similarities'' due to ambiguities of the French stenotypy method high rate of homophones, increased by ambiguities provided by the stenotypy method high rate of homographs, which cannot be efficiently reduced by a n-gram model

Stenotypy vs Speech recognition human interpretation  delete stammering and hesitations  speaker identification  add some extra speech events punctuations, breakpoints

specificities Very large vocabulary and specific lexicons New words in realtime Personnal adaptations independence of the stenotypy method

Personnal adaptations New stenotypy of a word Syntactic class modification Words concatenation post treatments

Ambiguity ? example 4 keystrokes - more than 360 paths - only 4 or 5 after linguistic analysis

Linguistic model Mixte approach : Linear combination of language models  3-gram  3-class knowledge rules

3 gram The 3-gram model is in fact a combination of 3-gram, 2-gram en 1-gram models Training corpus about 4.5 millions of words Lexicon about 150K most used words

3 class statistics association of Part Of Speech (POS)‏ tag set of 105 POS  NMS,  VA1PS,  XSOC, ...

Knowledge rules Used for special forms like  words after a 'apostrophe' must begin with by a vowel  The first word of the sentence must have capital  a verb after 'pour' is in infinitive form

Results NIST Scoring Toolkit (SCTK)‏ only a 5K words in test corpus manually corrected word error rate of 5% (1% earned this month)‏

Conclusion The first version of STENTOR is now out and is used by the profession World error rate already competitive Some improvements planned  long span dependencies  a better dictionnary  a larger training corpus

Conclusion Stentor a good lab but also a professional software with  Audio-sync  dictionaries builder  realtime word insertion  computer assisted correction  short cuts

Questions ? Thank you for your attention